Occluding contour detection and storage for digital photography

• US 20060204040A1
• Filed: 03/07/2005
• Published: 09/14/2006
• Est. Priority Date: 03/07/2005
• Status: Active Grant

• Current Assignee: Massachusetts Institute of Technology
• Original Assignee: Massachusetts Institute of Technology
• Inventors: Liu, Ce, Freeman, William T.
• Granted Patent: US 7,715,589 B2
• Time in Patent Office: Days
• US Class Current: 382/107
• CPC Class Codes: G06T 7/246 using feature-based methods, e.g. the tracking of corners or segments

Abstract

Described are techniques used in occluding contour detection. Cues obtained from a static image are combined with cues obtained based on motion disparity of objects in the image to determine occluding image contours. Two images of a same scene are obtained. Local pixel-wise flow fields are obtained using the Lucas-Kanade technique. A warped image is determined using the second image and the flow field. A disparity map is determined by comparing the warped image to the original first image. Static image cues are obtained from the original image using the Berkeley technique. The static image cues are combined with the motion disparity cues indicated by the disparity map producing a product image map. Occluding contours are determined by processing the product image map with the JetStream tracking technique. The occluding contours may be used with the original image for subsequent processing.

Claims

1. A method for determining occluding contours comprising:

receiving a first image and a second image of a same scene at different points in time;

determining motion disparity between said first image and said second image;

determining static image information about said first image; and

determining occluding contour information of said first image using said motion disparity and said static image information.

2. The method of claim 1, further comprising:

storing said occluding contour information with said first image.

3. The method of claim 2, wherein said occluding contour information includes a plurality of points corresponding to pixel locations of said first image.

4. The method of claim 2, wherein said occluding contour information includes an equation produced by a curve fitting process in accordance with a plurality of occluding contour points.

5. The method of claim 2, further comprising:

using said occluding contour information in at least one of;

image enhancement or image editing.

6. The method of claim 2, wherein said contour information is used in performing a non-interactive image operation on said first image.

7. The method of claim 2, wherein said contour information is used in performing an interactive image operation in said first image.

8. The method of claim 1, further comprising:

combining said static image information and said motion disparity into a product; and

tracking occluding contours of said product to produce said occluding contour information.

9. The method of claim 1, further comprising:

determining a flow field representing the motion disparity between said first image and said second image;

warping said second image in accordance with said flow field to a warped image; and

determining a disparity map by comparing said warped image with said first image.

10. The method of claim 1, wherein said static information includes at least one of:

color, texture and brightness.

11. The method of claim 8, further comprising:

determining a pixel in said product having a maximum intensity;

using said pixel as a starting point to track a contour;

removing said contour from said product producing a remaining product; and

repetitively determining other contours from said remaining product until a threshold intensity of said product is reached.

12. The method of claim 11, further comprising:

determining a plurality of paths from said starting point; and

determining said contour in accordance with an average of said plurality of paths.

13. The method of claim 1, further comprising:

receiving a second set of occluding contour information determined using a third image of said same scene;

determining final occluding contour information by combining said occluding contour with said second set of occluding contour information; and

storing said final occluding contour information with said first image.

14. A camera comprising:

a first component that receives a first image and a second image of a same scene at different points in time;

a second component that determines occluding contour information of said first image using motion disparity between said first image and said second image; and

a third component that stores said occluding contour information with said first image.

15. The camera of claim 14, wherein said occluding contour information is determined using static image information about said first image.

16. The camera of claim 15, further comprising:

a fourth component that uses said occluding contour information for performing an image enhancement.

17. A computer program product for determining occluding contours comprising code that:

receives a first image and a second image of a same scene at different points in time;

determines motion disparity between said first image and said second image;

determines static image information about said first image; and

determines occluding contour information of said first image using said motion disparity and said static image information.

18. The computer program product of claim 17, further comprising code that stores said occluding contour information with said first image.

19. The computer program product of claim 18, wherein said occluding contour information includes an equation produced by a curve fitting process in accordance with a plurality of occluding contour points.

20. The computer program product of claim 18, further comprising code that uses said occluding contour information in at least one of:

image enhancement or image editing.

21. The computer program product of claim 17, further comprising code that:

determines a flow field representing the motion disparity between said first image and said second image;

warps said second image in accordance with said flow field to a warped image; and

determines a disparity map by comparing said warped image with said first image.